The Farm Labor Problem · The Mover-Stayer Model With Heterogeneous Individuals 64 ... The UC...

The Farm LaborProblemA Global Perspective

J. Edward Taylor

Diane Charlton

Contents

Author Biography ixPreface xi

1. Introduction 1

The Problem of Farm Labor Demand 2The Agricultural Production Function 3Technological Change: Switching to a New Production Function 4Production Risk 5Timing and Seasonality 6Space 8Inequality and Concentration 10The Problem of Farm Labor Supply 10Equilibrium in Farm Labor Markets 12A Continuum From Family Farming to an Immigrant Farm

Workforce 13The End of Farm Labor Abundance 15Robots in the Fields 16References 17

2. Agricultural Labor Demand 21

Modeling Agricultural Labor Demand 22An Uncertain Relationship 24An Example: Lettuce Labor Demand on California’s

Central Coast 25Rethinking the Production Function 27A Two-Stage Crop Production Function 28Volatility in Farm Labor Demand 29Impacts of Yield and Wage Shocks on Lettuce Labor Demand 30Weather Shocks and Labor Demand in a Multistage Production Function30What Do the Data Show? 32Why Seasonality Matters So Much 34Appendix A 35Derivation of Farm Labor Demand 35Appendix B 42

v

A Multistage Crop Production Model 42Reference 46

3. The Farm Labor Supply: Who Does Farm Workand Who Does Not? 47

How Much to Work? 47Where to Work? 54The Mover-Stayer Model 54The Lewis Model: Too Much Labor on the Farm 57Migration and Unemployment: The Harris-Todaro Model 61The Mover-Stayer Model With Heterogeneous Individuals 64Harris and Todaro in a Mover-Stayer Model 64Applying the Mover-Stayer Model to Data 68Conclusion 71Appendix A 73Mathematical Derivation of Household Labor Supply 73Appendix B 74The Harris-Todaro Model in More Mathematical Detail 74References 76

4. Equilibrium and Immigration in the FarmLabor Market 77

A Spatial Equilibrium 79Immigration Changes Everything 83Follow-the-Crop Migration Redistributes Workers Within

Countries 86Linking Labor Supply With Demand: Farm Worker Recruitment 88Direct Hiring by Farmers 88Farmer Associations 88Unions and Hiring Halls 89Public Employment Services 89Farm Labor Contractors 90Farm Labor Recruiting in the Internet Age 90The Challenges of Seasonality for Workers and Communities 91Immigration and Farm Labor Across Countries 91Appendix A 93Modeling Farm Labor Market Equilibrium 93References 94

5. Labor in an Agricultural Household 97

Agricultural Households as Producers 100Agricultural Households as Consumers 101Putting Production and Consumption Together: An Autarkic

Chayanovian Farm Household 103Labor in an Agricultural Household Model With Well-Functioning

Markets 105

vi Contents

The Evolution of Agricultural Households 108Household Response to Price Changes 108Impacts of Rising Wages on an Agricultural Household 109Welfare Implications 110Migration, Education, and the Agricultural Transformation 110Conclusions 112Appendix 113Mathematically Modeling Production and Consumption

Decisions of Agricultural Households 113References 120

6. Farm Labor and Immigration Policy 121

Immigration and Farm Labor Market Equilibrium 123Immigrant Workers in US Agriculture 124From Railroad to Farm Worker 125Japanese Workers Replace the Chinese 128Dustbowl Migration Crowds Out Immigrant Labor in the Fields 130War Changes Everything 132The Second Bracero Program (1942–64) 132Illegal Immigration After the Bracero Program 134The 1986 Immigration Reform and Control Act 138Unintended Consequences: The Impacts of IRCA

on US Agriculture 141Importing Legal Farm Workers (at a Cost): The H-2A Program 143Efforts at Immigration and Farm Labor Reform After IRCA 144Guest Worker Programs in Other Countries 146Immigration and the Domestic Farm Labor Supply:

Chicken or Egg? 149Immigration Policy and Agriculture: Some Final Thoughts 151References 152Further Reading 154

7. Farm Labor Organizing From Cesar Chavezand the United Farm Workers to Fair Foods 155

The Plight of Farm Workers in America 156Cesar Chavez and the United Farm Workers Union 157The Decline of Farm Labor Unions 159How the “Salad Bowl Strike” Backfired 161Boycotts: A More Effective Bargaining Tool 163Farm Labor Unions Around the World 163The Fair Food Movement 164The Economics of Fair Food: Turning Market Power

on Its Head 166Fair Foods on a Country Scale? 171“Pricing In” the Ethical Treatment of Farmworkers 172Using Labels to Help Consumers Make More Informed

Decisions 173

Contents vii

Coordination of Consumers, Unions, and NGOs AcrossInternational Borders 174

Rich-Country Legislation to Protect Poor-Country Workers 175Exploiting Global Supply Chains 176Bananas and IFAs 176Flower Power 177Reasons for Hope 178References 179

8. The End of Farm Labor Abundance 181

When the Farm Labor Supply Becomes Less Elastic 184Was There a Shift in the Farm Labor Supply? A Quasinatural

Experiment 186How Quickly Is the Workforce From Rural Mexico Shifting

Out of Agricultural Work? An Econometric Analysis 187A Region-by-Region Look at the Declining Farm Labor Supply 192Why Does the Farm Labor Supply Become More Inelastic? 193Beyond Mexico and the United States: A Global Trend 197Farmers’ Options in an Era of Labor Scarcity 200The Exploration and Development Response 200The “Labor Conservation” Response 202References 203Further Reading 204

9. Robots in the Fields 205

The UC Tomato Harvester 206Modeling the Economics of Labor-Saving Technology Adoption 209Responses to a Shrinking Farm Labor Supply 214A Surge in Agricultural Innovation 215Conclusion 218Appendix 219Cost-Benefit Analysis of Adopting a Labor Saving Technology:

Raisin Grapes 219Economics of Technology Change in Labor-Intensive Raisin

Grape Production 220References 223Further Reading 225

Index 227

For more information, visit - https://farmlabor.ucdavis.edu/.

viii Contents

Chapter 1

Introduction

The fight is never about grapes or lettuce. It is always about people.

Cesar Chavez

Agriculture is different from other economic sectors of the economy in ways

that have far-reaching implications for the analysis of labor markets

(Timmer, 1988). A multitude of farms are scattered across a vast geographic

space. Production and labor demands are seasonal and uncertain, separated

in time and dependent on the whims of nature. Farmers make their planting

decisions months (and in the case of perennial crops, years) ahead of harvest.

Labor and other inputs have to be available exactly when farmers need them,

sometimes with little to no advance notice. Perishable crops not harvested on

time can rot in the fields. Late harvests can cause farmers to miss key marketing

windows or contract deadlines, and crops left in the field are exposed to disease

and weather risks. If labor is not available to plant crops and apply inputs on

time, or if the rains do not come, there might not be anything to harvest at

all. All of these considerations make the demand for farm labor uncertain

and dynamic—that is, it changes over time.

The supply of farm workers is also uncertain and dynamic. As countriesdevelop and their per-capita incomes rise, their workforces move out of agricul-

ture and into nonfarm jobs, creating the specter of farm labor scarcity. Farmers

in rich countries pressure their governments to open the doors to workers from

poorer countries to fill the void. Even if there is an overall abundance of farm

workers, local labor shortages materialize if workers are not available in the

right place at the right time. Crops ripen unevenly across space, creating

demand for follow-the-crop migrant workers. History shows that people do

not choose to do follow-the-crop migration if they have other options.

This chapter introduces the complex challenges associated with farm labor

demand and supply, the uncertain and seasonal equilibrium between the two,

and the role of immigration in addressing the farm labor problem. It concludes

with a view toward the future, discussing how agriculture must adjust to a new

farm labor equilibrium as people move off the farm and the global demand for

food continues to rise.

The Farm Labor Problem. https://doi.org/10.1016/B978-0-12-816409-9.00001-X

© 2019 Elsevier Inc. All rights reserved. 1

THE PROBLEM OF FARM LABOR DEMAND

The demand for farm labor derives from farmers’ production decisions, which

we will learn about in detail in Chapter 2. Agriculture differs from other indus-

tries in intrinsic ways that differentiate the timing and magnitude of labor

demand. The agricultural production process is biological. It relies heavily

on inputs from nature (land and weather). Consequently, agricultural production

is highly seasonal. There are long time lags between applying inputs and har-

vesting outputs. Although farmers can directly control how much they plant,

uncertainty surrounds how large the harvestable crop will be and how the har-

vested crop will be valued on the market. Farming requires land, so agriculture

is dispersed over a wide geographic area. Because agricultural production is

spread out geographically, involves long time lags, and is highly seasonal

and uncertain, timely access to labor, like other inputs, is critical to the success

and competitiveness of farm operations.

Agricultural production and access to labor vary substantially around the

world. In parts of the United States, large agribusinesses [which the American

journalist CareyMcWilliams referred to as “factories in the field”] dominate the

farm landscape (McWilliams, 2000). Millions of small family farmers dominate

production in low-income countries.1 In most of the world, households, not

firms, make most of the agricultural production decisions, and the family pro-

vides most or all of the labor needed on the farm. Often, hired labor is an imper-

fect substitute for family labor, because hired workers might not work as hard,

they might not be available for hire near the farm, and poor farmers might lack

the cash to hire workers, particularly in the preharvest period prior to receiving

payment for the harvest.

Agriculture is marked by a high degree of uncertainty; thus, considerations

of risk are a hallmark of agricultural decision-making (Moschini and Hennessy,

2001). There are two broad categories of risk in agriculture: Production risk and

marketing risk. On the plus side, nature provides inputs like sunshine and rain-

fall at no cost. The downside is that farmers cannot predict when the rains will or

will not come, whether there will be a long season with no sunshine, or whether

a swarm of locusts will devastate the crop. Shocks of nature break the engineer-

ing relationship between inputs and outputs. Variables outside farmers’ control

determine how much of a harvestable crop there will be on the tree or in the

field—not only the vagaries of weather, but also risks associated with pests

and access to inputs, including labor.

This sort of uncertainty generally does not arise in manufacturing, where

engineering relationships govern production processes. Farmers’ access to hired

workers and other inputs, credit, insurance, and a market for the harvested crop

1. There are an estimated 570 million farms in the world. A total of 35% are located in China and

24% in India. Small farms (less than 2 ha) constitute about 12% of farm operations worldwide, and

family farms constitute about 75% (Lowder et al., 2016).

2 The Farm Labor Problem

impacts how they will respond to seasonality and uncertainty in agricultural

production. New evidence suggests that climate change is increasing agricul-

tural production risk as well as impacting agricultural labor markets.2

Once the harvestable stock of produce is mature in the fields or on the trees,

agricultural production is largely a resource-extraction problem—how to harvest

the crop and get it to market in the most economically efficient way. Most of the

risk at this stage revolves around the availability of harvest labor and market

prices. Seasonal variations in spot-market prices create critical timing windows

for agricultural producers. Increasingly, grower-shippers have time-sensitive con-

tractual commitments as preferred suppliers to mass merchandisers (e.g., Wal-

mart and Costco), supermarket chains, and food service industries. This is true

in developing as well as high-income countries (Reardon et al., 2003). Failure

to harvest a field on time can result in the grower-shipper failing to meet delivery

commitments to buyers under Vender Managed Inventory Replacement (VMIR)

and other preferred supplier agreements. Some research suggests that, with the

consolidation of the retail sector, shippers have less negotiating power and are

more fearful of losing accounts if they fail to complywith buyer requests. Increas-

ing trade integration creates price competition and narrows marketing windows,

intensifying pressure on farmers. Having access to workers at critical moments

in the production process can make the difference between meeting delivery

commitments or not.

Labor supply risks are paramount at harvest time. An important potential

component of risk is the lack of available labor at the times and places needed

to harvest crops. Production risk can result from an insufficient labor supply if

fruit spoils on the trees before it can be harvested or if labor shortages prevent

farmers from marketing their harvest on time and complying with their contrac-

tual obligations higher up on the supply chain.

The Agricultural Production Function

The production function occupies center stage in farmers’ decisions, as it does

in the microeconomics of all firms. It describes a technological relationship, a

recipe to convert inputs into outputs. In most sectors of the economy, the pro-

duction function represents a known engineering relationship between inputs

and outputs, like how many copies of this book can be produced from a given

amount of paper, ink, capital (printing machines), labor, and so on. Like a

kitchen recipe, a production function can describe a fixed relationship

between inputs and outputs. For example, a tomato harvester might pick an

average of 30–33 1-pound buckets in an hour, implying around 1.9min of

2. Examples include (Mendelsohn et al., 1994; Schlenker and Roberts, 2009; Deschenes and

Greenstone, 2007; Jessoe et al., 2016).

Introduction Chapter 1 3

labor per harvested bucket.3 Most production functions are not linear, though.

They reflect decreasing marginal productivity of inputs as well as the possibility

that inputs may substitute for one another (unlike flour and salt in a pancake

recipe). Increased use of machinery can reduce labor needs. Even in the harvest,

diminishing returns eventually set in if too many workers are added to the

harvest crew.

The harvest depends on the stock of produce in the orchard or field ready to

harvest as well as the inputs applied at harvest time. The harvestable stock, in

turn, is uncertain, and it depends on decisions made prior to the harvest. Input-

output relationships involved in putting a harvestable crop in the field or on the

tree are different from those involved at harvest time. Agricultural production is

sequential, with different production functions describing input-output relation-

ships at different stages of the production process. The sequential nature of crop

production and risk means that we should not treat farm labor like other produc-

tion inputs or like labor in other sectors of the economy using a conventional

single-stage production model—even though most researchers do just that.

Commercial farmers, like other producers, use the technology at their dis-

posal to combine inputs and produce output, with the objective of maximizing

profits. (They may have other objectives as well, but for most purposes, profits

are a reasonable focus in agricultural models.) Profit maximization implies

employing additional variable inputs, including labor, up to the point where

the benefit of adding an additional unit of input (the marginal value product

of the input) just equals the input’s per-unit price, given the production function.

A farmer will not pay employees for an additional day of work unless this results

in a sufficient increase in gross revenue to cover the wage bill. If by hiring an

additional worker-day of labor the farmer will gain an increase in harvest worth

more than the wages paid, she will do it.4

Technological Change: Switching to a New Production Function

Over time, producers may invest in new labor-saving technologies to reduce the

costs of hiring workers. This implies switching to a new production function.

When mechanical harvesters (or more recently, robots) accompany humans

in the fields at harvest time, the recipe changes: fewer worker-days are required

per pound of harvested crop. The history of agricultural production is largely

about the development and adoption of new technologies that change the

input-output relationship. Labor-saving technologies, like the cotton gin and

combine harvesters, dramatically reduced labor needs in cotton and grain fields.

3. These tomato harvest labor figures come from a study entitled “Labor Requirements and Costs

for Harvesting Tomatoes,” by Zhengfei Guan, Feng Wu, and Steven Sargent, University of Florida;

http://gcrec.ifas.ufl.edu/media/gcrecifasufledu/docs/zhengfei/FE-1026—Tomato-Harvest.pdf.

4. One worker-day is a single worker employed for one day, two workers employed for½ day each,etc.


New IT-intensive technologies use robotics to reduce labor needs for tradition-

ally labor-intensive fruit and vegetable crops. The Green Revolution is another

example of technology change. New high-yielding seeds dramatically increased

yields the world’s major food crops, but Green Revolution technologies are land

saving, not labor saving: They enable farmers to produce a larger output on the

same amount of land. In most cases, they require additional labor to apply fer-

tilizer and other inputs as well as to bring in a larger harvest.

Adopting a new technology is costly. Farmers will only invest in a new tech-

nology if it passes the cost-benefit test (Chapter 9). The discounted benefits

(cost savings) from the investment over time must be greater than the cost of

the initial investment in order for a technology to be economically feasible.

The cost savings depend on the nature of the technology and also on input costs.

If the cost of an input (say, farmworker wages) is increasing over time, a labor-

saving technology becomes more attractive. If, on the other hand, land is scarce,

land rents are rising, and labor is abundant, a land-saving technology is more

likely to pass the cost-benefit test.

Farmers are like any commercial producer in that one of their goals is to

maximize profits. However, the differences between agriculture and other pro-

duction sectors have profound implications when it comes to determining the

optimal amount and timing of labor inputs with a given technology, or deciding

whether it is optimal to switch to a new technology.

Production Risk

The agricultural production process is biological and filled with uncertainty.

Thus, agricultural economists pay a great deal of attention to incorporating risk

into the crop production function. In agriculture, the production function is ran-

dom or stochastic [from the Greek word στόχος (stóchos), meaning “guess” or“target”]. We will learn about stochastic production analysis and what it means

for labor demand in later chapters, but for now it is important to bear in mind

this crucial difference between agriculture and other sectors. The amount of

inputs farmers demand with a given technology depends on yield and price

uncertainty.

The choice of technology also depends on risk. Often, there is a tradeoff

between risk and expected returns from new technologies. A high-yielding seed

might require greater investment in fertilizer and other inputs—not to mention

labor to apply the inputs—in return for a yield that is higher on average but more

variable than yields from conventional seeds. Yield risk is particularly a con-

cern on the marginal, rain-fed lands most of the world’s farmers sow. If the rains

do not come (or too much rain comes, or wind, or too much heat or cold),

farmers risk losing the crop as well as the money they spend on inputs ahead

of the harvest. It is not surprising, then, that marginal lands are where one is

least likely to find high-yielding crop varieties (HYVs) around the world.New technologies can also reduce risk. Agricultural research currently is

underway to develop seed varieties that are robust to drought and other weather


shocks, without necessarily increasing expected yield in normal years. In effect,

these new seeds come with their own built-in insurance policy.5 As labor

becomes scarcer and its availability becomes less certain, labor-saving crop

technologies can reduce production risks while making it possible to harvest

the same crop with fewer (and no doubt more expensive) workers. A farmer

might be willing to spend more on a labor-saving technology than the wage sav-

ings one expects to gain from it, provided that the technology makes one less

vulnerable to unexpected labor shortages at harvest time. The difference

between the adoption cost and expected wage savings could be justified as a

“harvest labor insurance premium.”

Timing and Seasonality

Farm labor employment is highly seasonal. A 1938 study of California farm

labor noted extreme fluctuations in seasonal labor demand (Barry, 2007):

“During the peak season of September of the year 1935...there were demands for

more than 198,000 workers, while in the slack season of the same year, during

December and January, but about 47,000 were required, leaving over 150,000

unemployed.”

More recent data from the US Bureau of Labor Statistics paint a similar picture.

Fig. 1.1 shows the seasonal ups and downs of the two components of the US

farm labor demand for fruit, vegetable, and horticultural production between

2001 and 2009: direct hiring by farmers and hiring by farm labor contractors

(FLCs), who supply workers to farms. For both, the peak-to-trough ratios

appear remarkably stable from one year to the next—and they are big.

Agriculture involves at least two production stages. In the preharvest season,

the farmer’s objective is to efficiently create a stock of harvestable fruit on the

tree (or vine, or field). In the harvest period, his objective is to turn the harvest-

able stock into a marketable product, picked and packaged. There is no reason to

think that the demand for labor looks at all the same in these two periods.

Take a grape farm, for example. It employs a few workers year round to irri-

gate, weed, prune, and perform other tasks required for the vine to produce fruit.

The harvest season, on the other hand, is short and intense. The Fresno, Cali-

fornia, raisin grape harvest used to be the most labor-intensive activity in US

agriculture, employing 40,000 to 50,000 workers for only a few weeks each fall

to pick the grapes and spread them out on paper trays to dry in the sun. Today,

many farms have switched to dry-on-the-vine technology, in which machines

shake the dried raisins into bins. This reduces the need for farm workers during

harvest dramatically, almost eliminating harvest labor risks.

A multistage production process means that agricultural labor demand tends

to be highly seasonal. Combine this with risk and we can begin to understand

5. For example, see Lybbert (2006).


200,000

300,000

400,000

500,000

600,000

700,000

800,000

2001 Apr Ju

l

Oct

2002 Apr Ju

l

Oct

2003 Apr Ju

l

Oct

2004 Apr Ju

l

Oct

2005 Apr Ju

l

Oct

2006 Apr Ju

l

Oct

2007 Apr Ju

l

Oct

2008 Apr Ju

l

Oct

United States

Farm labor contractors Fruits, vegetable, and horticultural production

FIG. 1.1 Employment by FVH producers and farm labor contractors in the United States is highly seasonal. (Data from Bureau of Labor Statistics Quarterly Censusof Employment and Wages (http://www.bls.gov/cew/data.htm).)

Intro

ductio

nChapter

17

why agriculture and farm labor markets are so different. Agricultural production

at each stage of the production process is stochastic. Nature’s surprises early inthe growing season can create large swings in labor demand come harvest time.

Whenever there are time lags between input use and harvest, producers have

to find ways to finance their inputs. They also have to find ways to put food on

the family table if the crop fails. Credit is vital to most farmers in high-income

countries; farmers can take out loans to purchase inputs, and then repay the

loans after harvest. These inputs include labor. Workers might be willing to

work on your farm, but only if you have the cash before harvest to pay them.

Crop insurance is a different matter. Most farmers who grow labor-intensive

fruit, vegetable, and horticultural (FVH) crops do not have an insurance policy

that will pay out if the crop fails. Grain farmers in the United States and Europe

do have access to crop insurance, but only because of generous government

programs.6

Credit markets are not available to most farmers in poor countries, and for-

mal crop insurance is virtually nonexistent outside of a few development pro-

jects.7 Yet poor farmers in developing countries, like farmers in rich countries,

take on a heavy risk burden when they buy inputs and hire workers prior to har-

vest, knowing that the crop might fail or the price of the harvested crop might

tumble.

Space

Farming requires land, so agriculture is spread over wide geographic areas.

Contrast this with, say, information and technology industries in California’s

Silicon Valley or Bangalore, India, which reap economies of agglomerationby locating close to one another so that they can share ideas and build on

one another’s innovations. Because farms are spread out, timely access to

diverse markets is crucial. This includes markets for the output farmers produce,

but it also includes markets for inputs—including labor.

Workers have to be available to farms in the right numbers and at just the

right times, or else crop losses are likely. A farmer might do everything right in

the preharvest periods, growing conditions might be ideal, and there might be a

bumper crop in the field, but if workers are not available to pick the crop at the

critical moment, the crop can rot before it is harvested. Even if labor shortages

leave, say, the last 15% of the crop unpicked, that might represent the farmer’s

profit margin.

6. You can learn about the Crop Insurance Program at the U.S. Department of Agriculture (USDA)

website: http://www.rma.usda.gov/aboutrma/what/history.html. The European Union has a less

comprehensive crop insurance program; e.g., see (Santeramo and Ford Ramsey, 2017).

7. An introduction to credit and insurance in developing countries appears in Taylor and Lybbert

(2015).


On a small family farm, the farmer’s family is likely to supply most or all of

the labor to cultivate and harvest crops. On many larger farms, including com-

mercial farms throughout the developed world, family labor is largely irrele-

vant. A large number of hired farm workers are needed for a short period of

time to bring in the harvest.

How do farmers find so many workers on short notice? In villages around

the world, information flows through networks of contacts between farmers and

workers and among workers. Sometimes landlords, worried about having suf-

ficient workers for harvest, enter into complex “labor tying” arrangements in

which they provide people with various benefits in return for a promise to work

on the landlord’s farm at times of peak labor demand. The benefits might

include one or a combination of employment during the off season (when peo-

ple have a hard time finding work), access to plots of land to farm, credit, or

inputs like seed and fertilizer.8

In John Steinbeck’s story of the great Dust Bowl migration, The Grapes ofWrath, desperate workers showed up at California farms begging for work atany wage. Many farmers rely on the same workers returning from one year to

the next to harvest crops. Some encourage this by providing workers with extra

hours of work, higher wages, and nonwage benefits; building personal relation-

ships with individual workers; or even by providing support to the communities

from which these workers come (in the US, some come from as far away as vil-

lages in southernMexico). The classic 1960 CBSNews documentary “Harvest of

Shame” followed migrant workers making their annual trek harvesting beans and

other crops up and down the US Atlantic Seaboard—often with children in tow.

Women play a key role in filling the harvest labor vacuum in many parts of

the world. In Chile, many women enter the workforce to harvest fresh fruits and

vegetables in the peak season, when wages increase, then return to their house-

holds after the harvests are in. In Morocco, women are far and away the main

source of labor to harvest and process the crocus flower stamen from which saf-

fron is produced. In Burkina Faso and many other African countries, women

manage their own small plots of land and supply most of the labor—often with

little or no help from men. Although tea harvesting machines exist (including a

Japanese version with advanced laser technology), most tea harvesting or

“plucking” is performed by women. A 2011 United Nations study found that

women comprised about 43% of the agricultural workforce globally and more

than 50% in some African countries (Jarvis and Vera-Toscano, 2004; Filipski

et al., 2017; Udry, 1996; SOFA Team and Doss, 2011).

Today in California, FLCs match thousands of farm workers with jobs on

many individual farms. Instead of hiring workers directly, a farmer can enter

into a contract with an FLC to, say, harvest 20 acres of oranges. The FLCs’ com-

parative advantage is their network of contacts with farm workers mobilized on

8. A classic work in this area is Bardhan (1984).


short notice and their intensive use of day labor markets, where workers gather

in the hope of getting a seat on a contractor’s bus.

Inequality and Concentration

Some sectors, like energy, steel, or automobiles, are dominated by a few large

producers. Agricultural production is different. It is by far the largest production

sector in the world in terms of numbers of producers. Of the world’s 570 million

farms, more than 475 million are less than 2ha in size (Lowder et al., 2014).

Thus, the vast majority of the world’s farms operate only a small share of

the world’s farmland. The world’s farmers have unequal access to resources,

from large agribusinesses with first-world infrastructure and market access to

small-scale farms producing for farmers’ markets to hundreds of millions of

poor family farmers struggling for subsistence in less-developed countries.

To understand the supply and demand for farm labor, we have to understand

the behavior of thousands, millions, or (in China and India) hundreds of millions

of heterogeneous actors. Influencing or simply understanding agricultural out-

comes requires having good economic models of diverse agricultural producers

and how they are likely to respond to different kinds of market, environmental,

or policy shocks. To complicate matters, as we shall see, what is beneficial for

large farmers may be detrimental for small farmers.

THE PROBLEM OF FARM LABOR SUPPLY

The farm labor supply comes from households, which decide howmuch of their

time to supply to farm and other work activities in order to maximize their wel-

fare, or utility (Chapter 3). In poor countries, agriculture is large relative to other

sectors in the economy, and agricultural labor is in abundant supply. Most coun-

tries start out rural and agricultural. In the early stages of economic develop-

ment, the farm labor supply is elastic (Lewis, 1954). That is, workers are

available to satisfy farm labor demands even at low wages, and they quickly

increase their labor supply when wages increase.

In high-income countries, agriculture is a small sector compared with indus-

try and services in terms of employment. The human capital (skill) requirements

for agriculture are minimal. One might expect, then, that the supply of labor to

agriculture, like any small sector, is elastic, easily expanding to meet seasonal

increases in labor demand. Contrast this with high human capital jobs, like com-

puter programming and health care, for which it might take several years to train

new workers when labor demands increase. Human capital investment (in par-

ticular, investment in schooling) is a major focus of dynamic labor supply

models. It is essential for doctors but not for farm workers.

Although it is possible to train farmworkers relatively quickly, people leave

farm work as economies develop and off-farm employment expands. This

results in a shrinking pool of domestic farm workers. In economic parlance,


the domestic agricultural labor supply shifts inward. People flee the farm. We

can imagine an equilibrium model in which agricultural wages increase apace

with nonagricultural wages, inducing domestic workers to stay on the farm

instead of moving to factory or service jobs. There might have to be an extra

wage premium to induce people to take farm jobs if nonfarm jobs bring other

nonpecuniary benefits, like being more interesting and less onerous than farm

work. Ask yourself: What would it take to get you out into the fields picking

crops on a hot summer afternoon? How long would you be willing to continue

doing field work?

History shows that rising farmwages do not induce large numbers of domes-

tic workers to do farm work. You will not find US-born unemployed steel

workers picking lettuce in California or oranges in Florida. The share of domes-

tic workers in the US hired farm work force has fallen to the point where, by

2006, only 23% of workers (2% in California) were US-born.9 The rest were

immigrants, earning a wage easily eight times what they could hope to earn

by working in rural Mexico, which is where most hired farmworkers in the

United States come from.

The wage rate in rural Mexico represents the opportunity cost of migrating

to US farm jobs.10 It is the lower bound on the reservation wage, the lowest

wage a worker would accept to migrate to the United States instead of working

in Mexico. A rural Mexican’s reservation wage also includes the economic and

psychic costs of migrating. Virtually all new immigrant farmworkers lack legal

status, so their reservation wage reflects the risk of being apprehended, jailed,

and returned to Mexico.11

The fact that most hired farmworkers in the US earn low wages and come

fromMexico suggests that the foreign farm labor supply is elastic. An elastic for-

eign labor supply at a relatively low wage can explain why the farm workforce in

the United States and other rich countries consists mostly of immigrants. The

quantity of domestic labor supplied to agriculture drops as nonfarm employment

expands, and foreign workers come in to fill the void. Any excess demand for

farmworkers gets filled by foreign workers, and this keeps wages lower than they

would bewithout immigration. In the United States, a growing reliance on foreign

agricultural workers and testimonials from farm workers, whose children usually

eschew farm work, support Martin’s contention that the farm workers of tomor-

row are growing up outside the United States (Martin, 2009).

Within farm labor-importing countries, the supply of labor to individual

farms and farm regions may be elastic in the short run, in the sense that a farm

9. Estimated from the U.S. Department of Labor’s NAWS Public Use Data (http://www.doleta.gov/

agworker/naws.cfm).

10. The market wage and marginal value product of labor are the same in a “separable” agricultural

household model; see Singh, Squire, and Strauss, 1986.

11. The migrant’s reservation wage may include other elements, as well: psychic costs of working

abroad, cost of living adjustments, etc.


can attract more workers by offering a slightly higher wage if it needs to. There

is an important spatial element to this story, though. Unless the total farm labor

supply can quickly adjust, farms can attract new workers only at the expense of

other farms. As fewer people in rural Mexico enter farmwork, farmwages in the

United States are rising. But as one observer put it:

The one constant is that no matter how much we pay, domestic workers are not

applying for these jobs. Raising wages only serves to cannibalize from the existing

workforce; it does nothing to add new laborers to the pool.

(Kitroeff, 2017)

The impacts of an excess demand for farm labor in one region reverberate into

other regions. A late harvest in one locality can prevent workers from migrating

to another locality. This produces the familiar pattern of localized seasonal farm

labor shortages, even if there is overall labor abundance and average farm

wages are low.

EQUILIBRIUM IN FARM LABOR MARKETS

Understanding how farm labor markets work requires integrating demand and

supply into farm labor market equilibrium models (Chapter 4). We use the

plural—“models”—here because there is no single farm labormarket anywhere.

Instead, there are many localized farm labor markets, often linked together by

migration.

In 2007, aWall Street Journal editorial claimed that “farmers nationwide arefacing their most serious labor shortage in years.” The editorial asserted that

“20% of American agricultural products were stranded at the farm gate” in

2006, including a third of North Carolina cucumbers, and it predicted that total

crop losses in California would hit 30% in 2007.

These predictions did not materialize. The example of cucumbers illustrates

that farm labor shortages tend to be local, not generalized across the nation. US

cucumber production rose both in 2006 and 2007 (Farm Labor Shortages,

Mechanization, 2007).

This and many other farm labor-shortage anecdotes suggest that, even if the

medium-to-long run supply of foreign workers is elastic, fed by new immigra-

tion, in the short run farmers compete across regions for a relatively inelastic, or

fixed, total supply of agricultural workers.

In an average year, what might be called a “first-order” equilibrium sup-

ported at the margin by follow-the-crop migration ensures ample labor at a pre-

dictable wage in a given place and season. However, within a network of

migration-linked farm labor markets, a stochastic shock to one local market

(e.g., a late pear harvest in the Sacramento Valley) reverberates into neighbor-

ing labor markets. One needs multiregion models to represent this spatial com-

plexity of farm labor supply and demand.


A number of researchers point out an important caveat to the seasonal equi-

librium just described. They note that the process of adjustment in farm labor

markets often is sluggish and incomplete. Studies focusing on developing coun-

tries argue that wage rigidities force farm labor markets to adjust through

changes in unemployment (Bardhan, 1979; Dreze and Mukherjee, 1989).

Jarvis and Vera-Toscano (2004), in a study of seasonal farm labor markets in

Chile, found that female workers absorbed most of the seasonal labor force

adjustments, because they had the lowest reservation wage.

In the United States, the availability of welfare and unemployment insur-

ance puts a lower bound on the reservation wage. Instead of migrating to follow

the harvest, many workers can stay put and receive unemployment insurance

and welfare support. Thus, farm labor markets are likely to adjust to demand

shocks via changes in the demand for welfare and unemployment insurance,

rather than through an endogenous wage adjustment that redistributes farm

workers across regions. Minimum wage laws reinforce this. A study by three

UC Davis researchers found that seasonal work, low earnings, and high unem-

ployment in California’s agricultural counties promoted welfare use and limited

the potential of local labor markets to absorb ex-welfare recipients. In Califor-

nia’s major agricultural counties, when unemployment peaks, welfare use

increases (Green et al., 2003).

A CONTINUUM FROM FAMILY FARMING TO AN IMMIGRANTFARM WORKFORCE

Perhaps the most notable difference between agriculture and other economic

sectors in poor countries is that most agricultural production is by households.

Most farms in the world are family farms that supply their own labor and other

inputs to the land and consume part or all of the harvest. Unlike firms that are

focused exclusively on production decisions, agricultural households make

both production and consumption decisions. This may seem like a technical dis-

tinction, but as we see in Chapter 5, it can drastically change the economic anal-

ysis of farm labor demand as well as supply.

Around the world, hired labor markets evolve over time in ways that mirror

the agricultural transformation. At low levels of economic development, family

farming dominates the agricultural landscape. Agricultural households provide

most of the agricultural inputs and consume a significant part (in most countries,

most) of the output from the farm. Not surprisingly, in predominantly subsis-

tence economies, agricultural labor markets are relatively “thin” and often

take the form of informal labor exchanges. In some areas, large plantations cre-

ate a demand for wage labor that coexists with subsistence production on

small farms.

The importance of hired labor expands as commercial production displaces

family production. Consider, for instance, the United States, where both family

and hired labor have declined over the last half century as a result of


mechanization, but the ratio of hired to family labor has increased, from 1:3 in

1950 to around 1:2 today (these averages mask large differences across US

regions) (Kandel, 2008). Increasing reliance on hired workers frequently gives

rise to internal migration to address labor market disequilibria across time and

space. However, a collection of factors—seasonality, uncertainty of farm

employment, aversion to follow-the-crop migration, the disagreeableness of

working conditions in the field, and expanding income opportunities outside

of agriculture—cause the domestic supply of agricultural labor eventually to

shift inward and upward. This creates incentives for farmers to seek a less

expensive source of labor abroad.

Migration policies evolve in imperfect harmony with farm labor market

trends, because they are the outcome of a complex and frequently internally

contradictory political process (Chapter 6). Rarely do countries implement pol-

icies explicitly to influence internal migration. Nevertheless, at the mid-stage of

the agricultural transformation, there tends to be growing concern for the wel-

fare of migrant farm laborers, their families, and the communities in which they

live. In some places, most notably California, farm labor organizing efforts

escalate (Chapter 7). At this stage, as Emerson (1989) states, “a major emphasis

of governmental policy toward migratory farm labor is to shift the migrant out

of the migratory stream, and if at all possible, to shift him to the nonfarm

sector.”12

This gives rise to an integration-immigration dilemma: In the absence of

labor-saving technological innovations and/or a shift to crops requiring less

labor, mobility out of agriculture for some workers implies the rotation of

new immigrants into the farm workforce. Martin (2009) describes an agricul-

tural “immigration treadmill” that perpetuates rural poverty by attracting a con-

tinual flow of new unskilled immigrants to take the place of workers who move

from farm to nonfarm jobs.

In high-income economies, agriculture loses its importance as a generator of

employment, while its dependence on foreign workers invariably increases. The

political process becomes a battleground in which farm interests engage the

interests of other actors, some of whom oppose the use of immigration policies

to guarantee an abundant supply of agricultural labor.

If border enforcement restricts immigration, the farm labor supply decreases

and wages increase. Meanwhile, the juxtaposition of high farmworker wages in

high-income countries with low earnings abroad keep the supply of immigrant

labor elastic, willing to cross borders to fill farm jobs. This intensifies pressure

at the border and makes immigration policies difficult to enforce. According

to conservative estimates, more than 50% of all US hired farmworkers are

12. Examples of government programs include the migrant and seasonal farm worker programs

launched as part of the War on Poverty in the 1960s, including Migrant Headstart, Migrant Educa-

tion, Migrant Health, and Job Training for Seasonal and Migrant Workers (Emerson, 1989).


unauthorized immigrants.13 The unauthorized share of workforces in other

high-income countries are not as well documented but certainly large.

Instead of seeking legal access to farmworkers from abroad, farmers could

adopt labor-saving technologies, invest in improved labor management prac-

tices, or take steps to evade immigration laws. There is evidence that penalties

against knowingly hiring unauthorized immigrants, which were included in the

United States 1986 Immigration Reform and Control Act (IRCA), accelerated a

shift away from direct hiring by farmers in favor of labor intermediaries, who

are more difficult for immigration law enforcement authorities to monitor

(Taylor and Thilmany, 1993).

THE END OF FARM LABOR ABUNDANCE

A pervasive theme in this chapter is that domestic agricultural workers become

increasingly scarce as countries’ incomes rise. Both logic and emerging empir-

ical evidence suggest that the same applies to the supply of foreign agricultural

labor. Labor-intensive agriculture in high-income countries seeks out new

migrant-source areas over time. What happens when foreign agricultural

workers move out of farm work? The evolution of agricultural labor markets

in farm labor-exporting regions raises questions about the sustainability of a

labor-intensive agricultural system dependent on immigrant labor, and indeed

of labor-intensive agricultural systems generally (Chapter 8).

Nowhere are agricultural labor markets more integrated across borders than

in Mexico and the United States. New research finds that the supply of farm

labor from rural Mexico is diminishing over time (Taylor et al., 2012;

Charlton and Taylor, 2016). United States and Mexican farmers compete for

an ever-smaller number of farm workers from rural Mexico. Migration is net-

work driven; newmigrants tend to follow past migrants to the same destinations

and jobs (Sana and Massey, 2005; Mora and Taylor, 2006; Pfeiffer et al., 2007).

As immigrant workers move out of agriculture and into factory and service jobs,

future migrants do, too. The supply of labor to agriculture diminishes.

In Mexico, the regional shift in labor supply to US farms is happening

quickly. Mexico has entered into agreements with Guatemala to import Guate-

malan farmworkers.14 Thus, Mexico is in a transitional phase of being both an

importer and exporter of farm workers. With a combined population one-third

the size of Mexico’s and accelerating urbanization, Central America’s potential

13. U.S. Department of Agriculture, Farm Labor. http://www.ers.usda.gov/topics/farm-economy/

farm-labor/background.aspx#legalstatus.

14. M�exico Centro de Estudios Migratorios/Instituto Nacional de Migración, Trabajadores guate-

maltecos documentados con la Forma Migratoria de Visitante Agrı́cola (FMVA) en el estado de

Chiapas; http://www.gobernacion.gob.mx/work/models/SEGOB/Resource/2796/1/images/Dossier

%206_%20Trabajadores%20guatemaltecos%20documentados%20con%20la%20forma%

20migratoria%20de%20visitante%20agr%c3%83%c2%adcola%20(FMVA)%20en%20el%

20estado%20de%20Chiapas.pdf


to replace Mexico as a significant source of US farm labor is limited. A 2017

headline in a major Mexican newspaper declared: “In agriculture, Mexico is left

without young producers (Georgina Saldierna, 2017).”

In Western Europe, immigrants from African nations work the fields of

Spain, France, and Italy; Albanians harvest crops in Greece; Polish workers toil

the fields in Germany and the United Kingdom. Regional integration under the

European Union (EU), including free labor movement among EU countries

under the Schengen Accord, boosted farm labor migration from the former

Soviet bloc countries. In time, one would expect that the migration networks

currently channeling eastern European workers into agricultural jobs in the

EU will become more urbanized, like networks from Mexico to the United

States.

Worried about its heavy reliance on foreign agricultural workers, Germany

offered farmers generous subsidies to hire German-born farm workers. That

experiment failed, though, and the program was abandoned (Txortzis, 2006).

Israel offered a different alternative to its farmers. In December 2009 it

announced a new public initiative to invest in mechanization of Israeli farming

“in an effort to reduce the need for foreign workers (Arutz Sheva, 2009).”

ROBOTS IN THE FIELDS

Countries dependent on imported agricultural labor have two options. The first

is to seek out new sources of farmworkers and (through immigration policy)

grant farmers access to them. This is not a long-run solution, though, because

the same process that shifts workers out of agriculture as economies grow and

contentious political processes limit rich countries’ ability to find—and

import—new sources of cheap farm labor over time.

The second option is to invest in reducing farm labor demands through a

combination of technological change, improved labor management practices,

and trade. Instead of large crews of low-skilled and poorly paid workers toiling

in the fields, this second option implies better-paid, skilled workers accompa-

nying robots in the fields (Chapter 9).

Stepping into a Wellsian time machine and glimpsing the agricultural future

in today’s farm labor-importing countries like the United States, we are sure to

find a combination of higher capital intensity and labor productivity, big labor-

saving innovations involving machine learning and robotics, a shift to more

imports of crops that are hard to mechanize, and higher farm wages.15 The farm

workforce will change; tekked-up workers will be needed to work in a tekked-

up agriculture.

The transition to this new agricultural world is likely to proceed unevenly

across countries, commodities, and agricultural tasks. The profitability of

labor-saving technological changes and crop mixes at a particular place and

15. The Time Machine is a science fiction novel by H.G. Wells.


time depends on the availability of low-cost labor. Policies facilitating access to

low-wage immigrant labor and developments abroad that stimulate emigration

will continue to discourage the adoption of labor-saving technologies. Mechan-

ical harvesters and new IT-assisted labor-saving solutions exist or are under

development for most crops, their usage limited mostly by low wages for har-

vest workers and quality and productivity concerns that can be addressed by

future research and development.

New developments in information and technology, machine learning, and

artificial intelligence point to a future of robot-assisted agricultural production.

Engineering departments at universities and high-tech startup firms in Silicon

Valley are beginning to develop labor-saving solutions for a diversity of crops

that combine mechanical engineering with artificial intelligence and machine

learning. The perishable nature of FVH crops, imperfect substitutability

between capital and labor, and consumer demands for high-quality locally

grown produce will insure that some production of labor intensive crops persists

even in high-wage countries. In time, though, the farm labor-immigration policy

connection is bound to weaken, as changes in the availability of low-cost labor,

technologies, crop mixes, and trade reduce countries’ reliance on imported agri-

cultural workers.

REFERENCES

Bardhan, P.K., 1979. Wages and unemployment in a poor agrarian economy: a theoretical and

empirical analysis. J. Politic. Econ. 87 (3), 479–500.

Bardhan, P.K., 1984. Land, Labor, and Rural Poverty. Cambridge University Press, Cambridge.

Barry, 2007. Influence of Employment Agencies on Migratory Farm Labor. p. 4.

Charlton, D., Taylor, J.E., 2016. A declining farm workforce: analysis of panel data from rural Mex-

ico. Am. J. Agric. Econ. 98 (4), 1158–1180.

Deschenes, O., Greenstone, M., 2007. The economic impacts of climate change: evidence from agri-

cultural output and random fluctuations in weather. Am. Econ. Rev. 97 (1), 354–385.

Dreze, J.P., Mukherjee, A., 1989. Labour contracts in rural India: theories and evidence. In: The

Balance Between Industry and Agriculture in Economic Development. Palgrave Macmillan,

London. pp. 233–265.

Emerson, R.D., 1989. Migratory labor and agriculture. Am. J. Agric. Econ. 71 (3), 617–629.

Farm Labor Shortages, Mechanization, 2007. Rural Migration News. vol. 13. No. 4 (October 2007).

Cucumber production Statistics are from USDA, http://www.nass.usda.gov/QuickStats/

index2.jsp.

Filipski, M., Aboudrare, A., Lybbert, T.J., Taylor, J.E., 2017. Spice price spikes: simulating impacts

of saffron price volatility in a gendered local economy-wide model. World Dev. 91, 84–99.

Green, R., Martin, P., Taylor, J.E., 2003. Welfare reform in agricultural California. Agric. Resour.

Econ. Rev. 169–183.

Jarvis, L., Vera-Toscano, E., 2004. Seasonal adjustment in a market for female agricultural workers.

Am. J. Agric. Econ. 86 (1), 254–266.

Jessoe, K., Manning, D., Taylor, E.J., 2016. Climate change and labour allocation in rural Mexico:

evidence from annual fluctuations in weather. Econ. J. https://doi.org/10.1111/ecoj.12448/

fullhttp://onlinelibrary.wiley.com.


Kandel, W., 2008. Profile of Hired Farmworkers, a 2008 Update (Economic Research Report Num-

ber 60) [Electronic Version]. Economic Research Service, United States Department of Agri-

culture, Washington, DC.http://digitalcommons.ilr.cornell.edu/key_workplace/559/.

Kitroeff, N., 2017. How this garlic farm went from a labor shortage to over 150 people on its

applicant waitlist. Los Angeles Times. February 9, 2017, http://www.latimes.com/business/

la-fi-garlic-labor-shortage-20170207-story.html.

Lewis, W.A., 1954. Economic development with unlimited supplies of labour. Manch. Sch. 22 (2),

139–191.

Lowder, S.K., Skoet, J., Singh, S., 2014. What do we really know about the number and distri-

bution of farms and family farms worldwide? Background paper for The State of Food

and Agriculture 2014, https://ilcasia.wordpress.com/2014/07/31/what-do-we-really-know-

about-the-number-and-distribution-of-farms-and-family-farms-in-the-world/.

Lowder, S.K., Skoet, J., Raney, T., 2016. The number, size, and distribution of farms, smallholder

farms, and family farms worldwide. World Dev. 87, 16–29.

Lybbert, T.J., 2006. Indian farmers’ valuation of yield distributions: will poor farmers value ‘pro-

poor’ seeds? Food Policy 31 (5), 415–441.

Martin, P.L., 2009. Importing Poverty?: Immigration and the Changing Face of Rural America. Yale

University Press.

McWilliams, C., 2000. Factories in the Field: The Story of Migratory Farm Labor in California.

University of California Press, Oakland.

Mendelsohn, R., Nordhaus, W., Shaw, D., 1994. The impact of global warming on agriculture: a

Ricardian analysis. Am. Econ. Rev. 84 (4), 753–771.

Mora, J., Taylor, J.E., 2006. Determinants of migration, destination, and sector choice: disentangling

individual, household, and community effects. In: InternationalMigration, Remittances, and the

Brain Drain. pp. 21–52.

Moschini, G., Hennessy, D.A., 2001. Uncertainty, risk aversion, and risk management for agricul-

tural producers. Handb. Agric. Econ. 1, 87–153.

Pfeiffer, L., Richter, S., Fletcher, P., Taylor, J.E., 2007. Gender in economic research on interna-

tional migration and its impacts: a critical review. In: The International Migration of Women.

pp. 11–51.

Reardon, T., Timmer, C.P., Barrett, C.B., Berdegu�e, J., 2003. The rise of supermarkets in Africa,

Asia, and Latin America. Am. J. Agric. Econ. 85 (5), 1140–1146.

Saldierna, G., 2017. En Materia Agrı́cola, M�exico Se Queda Sin Productores Jóvenes. La Jornada.

October 15, 2017, http://www.jornada.unam.mx/2017/10/15/sociedad/033n1soc.

Sana, M., Massey, D.S., 2005. Household composition, family migration, and community context:

migrant remittances in four countries. Soc. Sci. Q. 86 (2), 509–528.

Santeramo, F.G., Ford Ramsey, A., 2017. Crop insurance in the EU: Lessons and caution

from the US. EuroChoices. https://doi.org/10.1111/1746-692X.12154/fullhttp://onlinelibrary.

wiley.com.

Schlenker, W., Roberts, M., 2009. Nonlinear temperature effects indicate severe damages to US

crop yields under climate change. Proc. Natl. Acad. Sci. 106 (37), 15594–15598.

Sheva, A., 2009. Israel National News Briefs. In: NIS 625M to be Invested in Mechanization of

Agriculture. http://www.israelnationalnews.com/News/Flash.aspx/177179.

SOFA Team, Doss, C., 2011. The Role of Women in Agriculture. ESA Working Paper No. 11-02,

United Nations Food and Agricultural Organization.http://www.fao.org/docrep/013/am307e/

am307e00.pdf.

Taylor, J.E., Lybbert, T.J., 2015. Essentials of Development Economics. University of California

Press, Oakland, CA (Chapter 12).


Taylor, J.E., Thilmany, D., 1993. Worker turnover, farm labor contractors, and IRCA’s impact on

the California farm labor market. Am. J. Agric. Econ. 75 (2), 350–360.

Taylor, J.E., Charlton, D., Yúnez-Naude, A., 2012. The end of farm labor abundance. Appl. Econ.

Perspect. Policy 34 (4), 587–598.

Timmer, C.P., 1988. The agricultural transformation. Handb. Dev. Econ. 1, 275–331.

Txortzis, A., 2006. When Germans join migrant field hands, the harvest suffers. Christ. Sci. Monit.

https://www.csmonitor.com/2006/0524/p01s04-woeu.html.

Udry, C., 1996. Gender, agricultural production, and the theory of the household. J. Polit. Econ.

104 (5), 1010–1046.


The Farm Labor Problem · The Mover-Stayer Model With Heterogeneous Individuals 64 ... The UC...

Documents

Transcript of The Farm Labor Problem · The Mover-Stayer Model With Heterogeneous Individuals 64 ... The UC...